Comprehensive Analysis of Discontinuous Space Vector PWM Techniques for a Five-Phase Voltage Source Inverter

Transcription

1 International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Volume 1, Issue, July - ptemer 15, PP Comprehensive Analysis of Discontinuous Space Vector PWM Techniques for a Five-Phase Voltage Source Inverter Mohd. Arif Khan hool of Electrical & Electronics Engineering College of Engineering, ience &Technology Fiji National University, FIJI mdarif7@rediffmail.com Astract: This paper develops discontinuous space vector PWM (DPWM) technique for a five-phase voltage source inverter (VSI). Space vector model of a five-phase VSI shows that there exist 3 space vectors with three different lengths forming three concentric decagons. Application of outer most and middle set of space vectors to implement Space vector PWM yield nearly sinusoidal output. Thus the proposed DPWM utilises the same set of space vectors to implement the modulation techniques. Performance is evaluated in terms of total harmonic distortion and weighted total harmonic distortion in output phase voltages. A significant reduction in switching losses is oserved. The simulation results are provided to validate the concept. 1. INTRODUCTION Multi-phase motor drives have gained much popularity in recent years and a numer of research papers have een pulished. The main reason is the inherent advantages offered y multi-phase motors such as reduction in the amplitude and increase in the frequency of torque pulsation, reduction in the rotor current harmonics, reduction in the dc link current harmonics, reduction in the current per phase without increasing the voltage per phase leading and increasing the torque per ampere for the same volume machine. Keeping in view these advantages the application of multi-phase motors are coming up mainly in high power ranges such as ship propulsion, electric and hyrid vehicles, aircraft fuel pump applications etc. A review on multi-phase motor drives is availale in [1,]. Multi-phase motors need invarialy some sort of power electronic converter for their supply as phases more than three is not readily availale from the grid. The most common choice is a multi-phase voltage source inverter. There are mainly two methods of controlling the output voltage and frequency of inverters namely; square wave mode and pulse width modulation mode. A numer of PWM techniques are availale to control a three-phase VSI [3, 4]. However, Space Vector Pulse Width Modulation (SVPWM) has ecome the most popular method ecause of the easiness of digital implementation and etter DC us utilisation, when compared to the ramp-comparison sinusoidal PWM method. Another PWM method known as Discontinuous PWM is widely used ecause it offers reduced numer of switching and consequently reduced switching losses. This aspect ecomes extremely important when dealing with high power drive system, as even a small saving in switching losses means a large amount of overall power saving and thus enhanced energy efficiency of motors. In principle, there is a lot of flexiility availale in choosing the proper space vector comination for implementing space vector PWM for a multi-phase VSI. Space vector PWM technique for a fivephase VSI is illustrated in [5-1], where continuous mode is considered. This paper takes up issue of space vector PWM for a five-phase VSI in discontinuous mode. It is shown that the numer of switching and consequently the switching losses can e greatly reduced y tiding one or more inverter legs to either positive or negative DC us. This is termed as discontinuous space vector PWM which is developed for a five-phase VSI in this paper. Discontinuous space vector PWM is also availale for a five-phase VSI in [11] where natural extension of three-phase discontinuous PWM was done and only large length vectors were used. Although the maximum possile fundamental output is high equal to.6115 p.u. ut the output contain low-order harmonics. The switching is reduced y 4% compared to the continuous SVPWM. This paper utilises large and medium length space vectors to implement discontinuous SVPWM, providing sinusoidal output phase voltages. Similar discontinuous space vector PWM techniques are availale in [1]. However, paper does not provide detail of the common mode voltage and other aspect of the PWM. In contrast this paper illustrates leg voltages, common mode voltages and the amount of reduction in switching due to each scheme. Moreover, two ARC Page 1

2 Mohd. Arif Khan novel methods are proposed in this paper, which yield etter results compared to the existing schemes. The simulation results are provided to support the findings.. MODELING OF A FIVE-PHASE VSI Power circuit topology of a five-phase VSI is shown in Fig.1. The inverter input DC voltage is regarded further on as eing constant. The load is taken as star connected and the inverter output phase voltages are denoted in Fig.1 with lower case symol (a,,c,d,e), while the leg voltages have symols in capital letters (A,B,C,D,E). The model of five-phase VSI is developed in space vector form in [13], assuming an ideal commutation and zero forward voltage drop. A rief review is presented here..5v dc A B C D E a c d e Load Fig1. Power circuit of a Five-phase voltage source inverter There are ten switching devices and only five of them are independent, as the operation of two power switches of the same leg is complimentary. The comination of these five switching states gives out thirty two (3) space voltage vectors. Out of thirty two space vectors thirty are active vectors and two zero vectors. At any instant of time, the inverter can produce only one space vector. The relationship etween the machine s phase-to-neutral voltages and inverter leg voltages are given with v v v v v a c d e 4 5v 1 v v v v 5 A B 4 5v 1 v v v v 5 B A 5v 1 v v v v C A C C B D D D E E 4 5 (1) 4 5v 1 v v v v 5 D A 4 5v 1 v v v v 5 E A B B C C E E D where the inverter leg voltages take the value of ±.5 V DC. As noted, lower case letters in indices define phase-to-neutral voltages. Space vector of phase voltages defined, using power variant transformation, as given in [14]: v dq * ( v a av a v c a v d 5 * a v e ) () wherea = exp(j/5), a = exp(j4/5), a* = exp(-j/5), a* = exp(-j4/5) and * stands for a complex conjugate. The phase voltage space vectors thus otained in d-q plane are shown in Fig.. Since it is a five-phase system, transformation is further done to otain space vectors in x-y plane using equation (3) and the resulting space vectors are shown in Fig. 3. v xy 4 6 ( v a a v a v c a v d 5 8 a v e ) (3) n International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page

3 Comprehensive Analysis of Discontinuous Space Vector PWM Techniques for a Five-Phase Voltage Source Inverter q-axis /5 d-axis Fig. Phase voltage Space vector representation of all the thirty two states (zero vectors at origin) y-axis /5 x-axis Fig3. Phase voltage space vectors in x-y plane. It can e seen from Fig. that the outer decagon space vectors of the d-q plane map into the inner decagon of the x-y plane (Fig. 3), the innermost decagon of d-q plane forms the outer decagon of the x-y plane, while the middle decagon space vectors map into the same region. Further, it is oserved from the aove mapping that the phase sequence a,,c,d,e of the d-q plane corresponds to a,c,e,,d sequence of the x-y plane. 3. CONTINUOUS SPACE VECTOR PWM SCHEMES FOR A FIVE-PHASE VSI The purpose here is to generate sinusoidal output phase voltages using space vector PWM. Application of two neighouring medium active space vectors together with two large active space vectors in each switching period makes it possile to maintain zero average value in the second plane [5-1] and consequently providing sinusoidal output. Use of four active space vectors per switching period requires calculation of four application times or dwell times, laelled here T al, Tl, Tam, Tm. The expressions used for calculation of dwell times of various space vector are [5-6]; T al T a v l v l v m T am T a v m v l v m T l T v l v l v m T m v m T v l v m (4) To T s T al T am T l T m International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 3

4 Mohd. Arif Khan Large vector length is val vl vl V cos / 5 (5) 5 DC Corresponding medium vector length is v v v V (6) am m m Where T a * s l 5 DC v sin k / 5 Ts (7) v sin k / 5 T * s l v sin k 1 / 5 Ts (8) v sin k / 5 This method divides the total vector dwell times to their respective lengths. This is in essence allocates 61.8% more dwell times to large space vectors compared to medium space vectors thus satisfying the constraints of producing zero average voltage in the x-y plane. This can e more clearly seen from Fig. 4. q axis v phase t t s v v 1 phase * v s /5 v 11phase v 1phase d axis t t a s v a Fig4. Principle of calculation of vector application times (d-q plane) y axis v phase v 1phase / 5 v 11phase x axis v 1 phase Fig5. Principle of calculation of vector application times(x-y plane) International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 4

5 Comprehensive Analysis of Discontinuous Space Vector PWM Techniques for a Five-Phase Voltage Source Inverter It is seen from Fig. 4 that the vectors in x-y plane are in such a position to cancel each other y using the dwell time equations (4). The application of active and zero space vectors are arranged in such a way as to otain a symmetrical SVPWM. The space vector disposition in sector I is illustrated in Fig. 5. Switching pattern is a symmetrical PWM with two commutations per each inverter leg. The space vectors are applied in odd sectors using sequence v, v, v, v, v, v, v, v, v, v,,] [ al m am l 31 l am m al v, while the sequence is [ v, vl, vam, vm, val, v31, val, vm, vam, vl, v ] in even sectors. It can e easily oserved from equation (4) that the zero vector application time remains positive for * vs.557vdc. Thus the output phase voltage from a VSI using this Space vector PWM scheme is.557v dc. The sequence of vectors applied and corresponding switching pattern for sector 1 is shown in Fig. 6 where states of five inverter legs take values of -.5Vdc and +.5Vdc (referencing to mid-point of the dc supply is applied) and the five traces illustrate, from top to ottom, legs A,B,C,D and E, respectively. T/ T11/ T/ T1/ T1/ T31/ T31/ T1/ T1/ T/ T11/ T/ S V V11 V V1 V1 V31 V31 V1 V1 V V11 V Fig6. Switching waveform for sector 1 using large and medium vectors. 4. PROPOSED DISCONTINUOUS SPACE VECTOR PWM It is possile to move the position of the active voltage pulses within the half switching interval, to eliminate one zero output voltage pulse. Modulation strategies using this concept are termed as discontinuous modulation. This method causes one or more inverter leg to tie to either positive or negative dc us. This scheme is mainly important for high power applications where switching losses are considerale. Six different Discontinuous space vector PWM techniques are proposed and presented in this section. However, all the schemes essentially just rearrange the placement of the zero t is kept space voltage vectors within each half carrier or carrier interval. The method where 31 zero for the complete fundamental cycle is termed as DPWMMIN and the method where t is kept zero for one fundamental cycle is termed as DPWMMAX. Rest of the four methods splits the sectors and arranges the zero space vectors in various fashions are termed as DPWM, DPWM1, DPWM and DPWM3. The time of application of various space vectors are still governed y equation 4. The fundamental output voltage magnitude and the nature of phase voltage waveforms remains sinusoidal as that of continuous SVPWM. The switching waveforms for different discontinuous PWM methods are shown in Fig. 7. International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 5

6 Mohd. Arif Khan T/ T11/ T/ T1/ T1/ T1/ T1/ T/ T11/ T/ S V V11 V V1 V1 V1 V1 V V11 V Fig7a. Switching waveform for sector 1 using large and medium vectors for DPWMMIN S T13/ T/ T3/ T1/ T31/ T31/ T1/ T3/ T/ T13/ V13 V V3 V1 V31 V31 V1 V3 V V13 Fig7. Switching waveform for sector using large and medium vectors for DPWMMAX S T13/ T/ T3/ T1/ T31/ T31/ T1/ T3/ T/ T13/ V13 V4 V3 V14 V31 V31 V14 V3 V4 V13 (i) International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 6

7 Comprehensive Analysis of Discontinuous Space Vector PWM Techniques for a Five-Phase Voltage Source Inverter T/ T15/ T4/ T5/ T14/ T14/ T5/ T4/ T15/ T/ S V V15 V4 V5 V14 V14 V5 V4 V15 V (ii) Fig7c. Switching waveform for (i) sector 3 (ii) ctor 4 using large and medium vectors for DPWM. T/ T15/ T6/ T5/ T16/ T16/ T5/ T6/ T15/ T/ S V V15 V6 V5 V16 V16 V5 V6 V15 V (i) S T17/ T6/ T7/ T16/ T31/ T31/ T16/ T7/ T6/ T17/ V17 V6 V7 V16 V31 V31 V16 V7 V6 V17 (ii) Fig7d. Switching waveform for (i) sector 5 (ii) sector 6 using large and medium vectors for DPWM1. International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 7

8 Mohd. Arif Khan S T17/ T6/ T7/ T16/ T31/ T16/ T7/ T6/ T17/ V17 V6 V7 V16 V31 V16 V7 V6 V17 Fig7e. Switching waveform for sector 6 using large and medium vectors for DPWM. T17/ T8/ T7/ T18/ T31/ T18/ T7/ T8/ T17/ S V17 V8 V7 V18 V31 V18 V7 V8 V17 Fig7f. Switching waveform for sector 7 using large and medium vectors for DPWM3 It is seen from Fig. 7 that one leg is tied to either positive or negative dc us so no switching takes place in that leg. There are ten sectors and ten switches in a five-phase VSI thus one leg remains idle in two sectors i.e. for 7, upper switch of a leg is inoperative in one sector and lower one in other one sector. In DPWMMIN each leg is tied to the DC us for two susequent sectors. For instance in DPWM upper switch of leg D is tied to lower dc us in sector 1 and the upper switch of the same leg is tied to the upper dc us in sector 6. Thus the numer of switching is now reduced y one fifth compared to their continuous counterpart and consequently a corresponding reduction in switching losses. All the proposed schemes are simulated using Matla/Simulink and the resulting waveforms are shown in Fig 8. Part (a) of Fig 8 shows the placement of zero vectors in different sectors and part () of Fig 8 shows the waveforms for each schemes where V a leg (average leg voltage), V a (average phase voltage), V nn (average voltage etween neutral points) called common mode voltage. The switching frequency is taken as 5 khz and the fundamental frequency is chosen as 5 Hz. The reference input is varied from to the maximum otainale (.557 (p.u.)). International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 8

9 Comprehensive Analysis of Discontinuous Space Vector PWM Techniques for a Five-Phase Voltage Source Inverter β V4( 1 1 ) V3(1 1 1 ) V5( ) V14(1111) V13(1) V(1 1 ) V15(1) V1(1111) V6( 1 1 ) V16(1111) V11(1) V1(1 1 1) α V17(1) V(1111) V7( 111 ) V18(1111) V19(1) V1(1 1) V8( 1 1) V9(1 1 1) DPWMMIN β V4( 1 1 ) V3(1 1 1 ) V5( ) V14(1111) V13(1) V(1 1 ) V15(1) V1(1111) V6( 1 1 ) V16(1111) V11(1) V1(1 1 1) α V17(1) V(1111) V7( 111 ) V18(1111) V19(1) V1(1 1) V8( 1 1) V9(1 1 1) β DPWMMAX V4( 1 1 ) V3(1 1 1 ) V5( ) V14(1111) V13(1) V(1 1 ) V15(1) V1(1111) V6( 1 1 ) V16(1111) V11(1) V1(1 1 1) α V17(1) V(1111) V7( 111 ) V18(1111) V19(1) V1(1 1) V8( 1 1) V9(1 1 1) DPWM β V4( 1 1 ) V3(1 1 1 ) V14(1111) V5( ) V15(1) V13(1) V(1 1 ) V1(1111) V6( 1 1 ) V16(1111) V11(1) V1(1 1 1) α V17(1) V(1111) V7( 111 ) V18(1111) V19(1) V1(1 1) V8( 1 1) V9(1 1 1) DPWM 1 International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 9

10 Mohd. Arif Khan β V4( 1 1 ) V3(1 1 1 ) V5( ) V14(1111) V13(1) V(1 1 ) V15(1) V1(1111) V6( 1 1 ) V16(1111) V11(1) V1(1 1 1) α V17(1) V(1111) V7( 111 ) V18(1111) V19(1) V1(1 1) V8( 1 1) V9(1 1 1) DPWM β V4( 1 1 ) V3(1 1 1 ) V5( ) V15(1) V14(1111) V13(1) V1(1111) V(1 1 ) V6( 1 1 ) V16(1111) V11(1) V1(1 1 1) α V7( 111 ) V17(1) V18(1111) V(1111) V1(1 1) V19(1) V8( 1 1) V9(1 1 1) DPWM 3 Fig8. Wave forms for different DSVPWM schemes. Basic schemes designated as DPMWMMIN and DWPWMMAX tied each leg to negative dc rail and positive dc rail, respectively and thus the name MIN and Max are given to them. These schemes are thus not recommended for use as they may damage either the lower or upper switches of VSI. To avoid this situation three more techniques namely DPWM, DPWM1 and DPWM are suggested in which DPWMMIN and DPWMMAX are applied alternatively in each sector so that each leg can e kept inoperative alternately providing a symmetric switching. Each of the three methods proposed here are suitale to feed different types of loads. It can e oserved from Fig. 8, for DPWM that the discontinuous period is 18 lagging the phase voltage and thus the most suitale load will e the one operating at cos 18 power factor leading. This is ecause of the fact that in the range of discontinuous period the current is maximum and the numer of switching is minimum offering lower cos 18 lagging. It is further switching losses. Similarly for DPWM1 the suitale load power factor is seen for DPWM1 and DPWM the discontinuity is splitted in two parts and thus two different loads may e catered are cos 9,cos 4 power factor lagging and or leading. Simulation is once again carried out for different loading conditions. The load considered is standard series RLC type of load whose transfer function is given y equations (9) and the resulting average voltage, average current and leg voltages are shown in Fig. 9. (1/ L) s TransferFunction (9) s ( R / L) s 1/ LC Keeping R = 1 ohm and Xc = 1 ohm, ( ) s TransferFu nctionfor 9deg leading (9a) s ( ) s ( ) s TransferFu nctionfor 9deg lagging (9) s ( ) s International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 1

11 Inverter Phase 'a', leg 'a' voltages and current (p.u.) Inverter Phase 'a', leg 'a' voltages and current (p.u.) Inverter Phase 'a', leg 'a' voltages and current (p.u.) Comprehensive Analysis of Discontinuous Space Vector PWM Techniques for a Five-Phase Voltage Source Inverter ( ) s TransferFu nctionfor 18deg leading (9c) s ( ) s ( ) s TransferFu nctionfor 18deg lagging (9d) s ( ) s DPWM at 18 deg. lagging.4 Va. -. Ia -.4 Valeg Time(sec) Fig9a. Average leg and phase voltages and current for 18 degree lagging load for DPWM..8.6 Va DPWM1 at 18 deg. leading Ia Valeg Time(sec) Fig9. Average leg and phase voltages and current for 18 degree leading load for DPWM DPWM at 9 deg. lagging.4. Va Valeg Ia Time(sec) Fig9c. Average leg and phase voltages and current for 9 degree lagging load for DPWM. International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 11

12 Inverter Phase 'a', leg 'a' voltages and current (p.u.) Mohd. Arif Khan.8.6 Va DPWM3 at 9 deg. leading Ia Valeg -.6 Fig9d. Average leg and phase voltages and current for 9 degree leading load for DPWM3. It is seen from Fig. 9 that the load profile exactly matches the requirement of the modulator (current peak coincides with the discontinuous region of the leg voltages) and thus the switching losses in these cases are minimum. 5. GENERALIZED DISCONTINUOUS SVPWM A generalised discontinuous Space vector PWM was suggested in [15] for a three-phase VSI. The same concept is extended here for a five-phase VSI. A generalised neutral or common mode voltage is generated which is then injected into the reference voltage to otain a set of modulating signals. These modulating signals are then compared with high frequency triangular wave to generate the gate drive signals. Although this is a carrier-ased PWM method ut it produces the output of the similar quality to that of the space vector PWM. The method can e explained using Fig. 1. Modulator phase angle is denoted y and is measured from the intersection point of the two reference waveat t / 5. The common mode voltage shown as shaded portion in Fig. 1 is otained as; nn v, v, v, v, v *.5V maxv, v, v, v v v sgn, (1) a c d e dc a c At first the maximum reference voltage is identified and then the difference etween the availale dc us voltage.5v dc and the maximum of the reference yield common mode voltage. The control / 5 to keep the operation of modulator in the linear range. range of 6. PERFORMANCE EVALUATION Time(sec) Two parameters are taken in consideration for performance evaluation of the proposed DSVPWM methods namely Total Harmonic Distortion (THD) and Weighted Total Harmonic Distortion (WTHD) and are defined in [3] y equations (11) and (1), d e V THD = n n3,5,7.. V 1 (11) THD WTHD = ( V1 / 1 L = ) Vn n n V1 Where V n represents n th order harmonic component and V 1 represent fundamental output phase voltages. The lower order harmonic contents (upto 5 th order) are considered for calculation of THD and WTHD. The simulation is carried out to determine these performance indices for the complete range of the modulation index. The resulting THD and WTHD are shown in Fig. 11. (1) International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 1

13 WTHD THD Comprehensive Analysis of Discontinuous Space Vector PWM Techniques for a Five-Phase Voltage Source Inverter.5Vdc A XX / 5 B XX C XX D XX E XXXXXX XXXXXX XXXXXX XXXXXX XXXXXX XXXXXX X XXXXXX X Amplitude (p.u.) I II III IV V VI VII VIII IX X t -.5Vdc XXXXXXXXXX XX X XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX Angle in degrees VnN t Fig1. Generalised Discontinuous PWM method THD of Large and Medium Vectors DPWMMIN THD DPWMMAX THD DPWM THD DPWM1 THD DPWM THD DPWM3 THD Modulation Index WTHD using Large and Medium vectors WTHDMI N WTHDMA X WTHD WTHD 1 WTHD WTHD modulation Index Fig11. THD and WTHD for various modulation index. It is seen from the Fig. 11 that at lower modulation index the est method is DPWMAX while at high modulation index DPWM1 offers minimum THD and WTHD. To further illustrate the variation of THD and WTHD two plots are shown in Fig. 1, one at.3 modulation index and other at unity modulation index. At Modulation Index =.3 International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 13

14 THD and WTHD (% of Fundamental) THD and WTHD (% of Fundamental) Mohd. Arif Khan.5 THD WTHD DPWMMIN DPWMMAX DPWM DPWM1 DPWM DPWM3 At Modulation Index = THD WTHD DPWMMIN DPWMMAX DPWM DPWM1 DPWM DPWM3 Fig1. THD and WTHD at modulation index.3 and 1 for different DPWM method. It can further e seen from Fig. 1 that at low modulation index the THD and WTHD of DPWMMAX is lower than all the schemes and DPWM1 has highest THD. Since the DPWMMAX cannot e employed for implementation purposes, thus the next est scheme is DPWM. At high modulation index the value of THD and WTHD is smaller compared to the value at lower modulation index, which implies that the output voltage is very near to the sinusoidal. DPWM1 has lowest THD and DPWMand DPWM3 has highest THD at high modulation index. 7. EXPERIMENTAL INVESTIGATION A Five-phase voltage source inverter is developed using intelligent power module from VI Micro systems, Chennai. Texas Instrument DSP TMS3F81 is used as the processor to implement the control algorithm. Since this DSP may coded in C or C++, it is more user friendly and they have dedicated 16 hardware PINS to generate the desired PWM signals. The PWM circuits associated with compare units make it possile to generate upto eight PWM output channels (per Event Manger) with programmale dead and and polarity. This DSP is specifically meant for use in motor drive purposes and it can control upto 8-phase two-level inverter. The control code is written in C++ language in Code composer studio 3.3 which runs in a PC. The control signal generated y PC is transferred to the DSP oard through RS 3 cale connected in parallel printer port of the PC. The DSP oard is connected to the Power Module through dedicated control cale. The DSP interfacing circuit along with required A/D and D/A converter is uilt on the DSP oard itself procured from VI Micro systems. The complete experimental set up is shown in Fig. 13. International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 14

15 Comprehensive Analysis of Discontinuous Space Vector PWM Techniques for a Five-Phase Voltage Source Inverter Fig13. Five-phase experimental set up. Fig14. Switching pattern for the proposed PWM 8. CONCLUSION Fig15. Leg voltage for DPWMMIN The paper present a PWM technique termed as Discontinuous Space Vector PWM for a five-phase voltage source inverter. There exist two methods of SVPWM in a five-phase VSI, one utilises only large space vector set while the other utilises oth large and medium vector sets. This paper utilises large and medium vectors to synthesise the input reference in discontinuous mode. This method of PWM offers a reduction in overall numer of switching and consequently switching losses. International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 15

16 Mohd. Arif Khan DSVPWM utilising large and medium space vector is seen to reduce the numer of switching to 1/5 th and consequently the switching losses. Alternatively the inverter switching frequency can e enhanced keeping the same inverter losses. Six different schemes are proposed and presented. The analysis is done on the asis of two performance indices namely THD and WTHD. It can e concluded that the DPWMMAX provide lowest THD and WTHD for low modulation index. However, this method is not recommended for practical implementation as this may shorten the life of inverter. DPWM offers the next est result and thus it may e used for implementation. At high modulation index DPWM1 is recommended for use. A generalised discontinuous PWM method is proposed ased on triangle comparison method. This method is easy to implement, it will provide different DPWM schemes depending upon the modulation index. The viaility of the proposed schemes is validated using simulation results. ACKNOWLEDGMENT Authors fully acknowledge the support for this work provided y CSIR standard research grant no. (4)/7/EMR-II. REFERENCES [1] G.K.Singh; Multi-phase induction machine drive research a survey, Electric Power System Research, vol. 61,, pp [] M.Jones, E.Levi; A literature survey of state-of-the-art in multiphase ac drives, Proc. 37 th Int. Universities Power Eng. Conf. UPEC, Stafford, UK,, pp [3] G.D.Holmes, T.A.Lipo, Pulse Width Modulation for Power Converters - Principles and Practice, IEEE Press ries on Power Engineering, John Wiley and Sons, Piscataway, NJ, USA, 3. [4] M.P. Kazmierkowski, R. Krishnan and F. Blaajerg, Control in power electronics- selected prolems, Academic Press, California, USA,. [5] A.Iqal, E.Levi, Space vector modulation scheme for a five-phase voltage source inverter, Proc. European Power Electronics (EPE) Conf., Dresden, Germany, 5, CD-ROM paper no. 6.pdf. [6] A.Iqal, E.Levi, Space vector PWM techniques for sinusoidal output voltage generation with a five-phase voltage source inverter, Electric Power Components and Systems, 6, vol. 34 no. [7] H.A.Toliyat, M.M.Rahmian and T.A.Lipo, Analysis and modelling of five-phase converters for adjustale speed drive applications, Proc. 5 th European Conference on Power Electronics and Applications EPE,Brighton, UK, IEE Conf. Pu. No. 377, 1993, pp [8] R.Shi, H.A.Toliyat, Vector control of five-phase synchronous reluctance motor with space vector pulse width modulation (SVPWM) for minimum switching losses, Proc. IEEE Applied Power Elec. Conf. APEC, Dallas, Texas,, pp [9] H.A.Toliyat, R.Shi, H.Xu, DSP-ased vector control of five-phase synchronous reluctance motor, IEEE Industry Applications Society Annual Meeting IAS, Rome, Italy,, CD-ROM paper no. 4_5. [1] P.S.N.deSilva, J.E.Fletcher, B.W.Williams, Development of space vector modulation strategies for five-phase voltage source inverters, Proc. IEE Power Electronics, Machines and Drives Conf. PEMD, Edinurgh, UK, 4, pp [11] M.A. Khan and A. Iqal, Discontinuous space vector PWM for a five-phase VSI with higher dc us utilisation, Proc. IEEE INDICON 7, 6-8 pt, 7, Bangalore, CD-ROM paper. [1] X.F.Zhang, F.Yu, H.S.Li and Q.G. Song, A Novel Discontinuous Space Vector PWM Control for Multiphase Inverter Proc. Int. Symp. Power Electronics, Electrical Drives Automation and Motion SPEEDAM, Taormina, Italy, 6, CD-ROM paper S8-6. [13] A. Iqal,S. Moinuddin, Space vector model of a five-phase voltage source inverter, Proc. IEEE International Conf. On Industrial Technology (ICIT6), 15-17Dec. 6 Mumai, India, pp [14] E.E.Ward, H. Härer, Preliminary investigation of an inverter-fed 5-phase induction motor, Proc. IEEE Power Elec. Spec. Conf. PESC vol. 116, no. 6, 1969 pp [15] A. Hava, R. J. Kerkman and T.A. Lipo, A high performance generalised discontinuous PWM algorithm, IEEE Trans. Ind. Appl. Vol. 34, no. 5, sept/oct International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 16

17 Comprehensive Analysis of Discontinuous Space Vector PWM Techniques for a Five-Phase Voltage Source Inverter [16] Zakir Husain, R.K. Singh, S. N. Tiwari, Balancing of Unalanced Load and Power Factor Correction in Multi-phase (4-phase) Load Circuits using DSTATCOM, International Conference of Electronics and Electrical Engineering, Imperial College, London U.K, June 3-july, 1. [17] Zakir Husain, R.K. Singh and S.N. Tiwari, A Novel Algorithm for four-phase (Multi-phase) Source Side Load Balancing and Power Factor Correction", International Journal of Recent Trends in Engineering, Vol.1, No.3, pp , 9. AUTHOR S BIOGRAPHY Mohd. Arif Khanreceived his B.E. (Electrical) and M.Tech (Power System & Drives) degrees and PhDin 5, 7 and 1, respectively, from the Aligarh Muslim University, Aligarh, India. He has worked as nior Research Fellow in a CSIR project, Space Vector Pulse Width Modulation Techniques for Multiphase Voltage Source Inverter. Presently he is working with hool of electrical & electronics engineering, CEST, Fiji National University, Suva, Fiji as Assistant Professor. His principal research interest is Multi-phase Multi-motor machine drives system. International Journal of Research Studies in Electrical and Electronics Engineering (IJRSEEE) Page 17